Despite advances in immunosuppression, renal allograft survival remains limited to about 10 years. This is particularly important in pediatric patients due to their expected lifespans compared to adults (whose outcome most often is death with a functioning graft). In numerous studies, human cytomegalovirus (HCMV) infection has been associated with inferior transplant outcome which is improved by ganciclovir prophylaxis, but the mechanisms underlying these observations remain cryptic. The pathogenesis of CMV mediated allograft injury is important because viral allograft injury may be potentially modifiable to improve transplant survival. HCMV infection is associated with natural killer (NK) cell activation and acute rejection in renal transplant patients. NK cell activation itself also correlaes with acute and late rejection, but a direct association between HCMV, NK cells, and transplant outcome has not yet been defined. These findings are recapitulated in a nontolerant murine renal transplant model of rejection, with both NK infiltrates and acute rejection intensified in murine CMV (MCMV) infected allografts. Ganciclovir reduces NK infiltrates and improves graft damage in the murine model, resembling improved outcome in patients receiving ganciclovir prophylaxis. In the animal model, NK depletion reduces MCMV associated damage, supporting a direct role of NK cells in viral allograft injury. Thus, both human transplant populations and th murine renal transplant model exhibit remarkably similar findings supporting a pathogenic role of NK cells in CMV induced allograft injury. We will use this model to define mechanisms of MCMV induced NK activation in renal transplantation, followed by manipulations of these host and viral parameters of NK activation to modulate viral allograft injury. Such studies are essential to identify potential pathways that can be modified by clinical interventions to ameliorate viral allograft injury, yet cannot be performed in human subjects. However, given the limitations of animal models in precise recapitulation of human disease, we will expand the relevance of the animal model intragraft findings by analyzing NK activation in clinical allograft biopsies from a population of pediatric renal transplant recipients, then correlating the biopsy findings with HCMV infection, peripheral blood NK activation, and graft outcome in these patients. Understanding the pathogenesis of CMV associated allograft injury using both the animal model and a relevant patient population offers what we believe is the most promising approach for the development of strategies to improve transplant survival for HCMV infected patients.